The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references standard third-party Python libraries used for parsing document formats.
Evidence: Documentation in SKILL.md and README.md suggests installing pdfplumber, python-docx, and PyPDF2 to handle PDF and Word file extraction.
[COMMAND_EXECUTION]: The skill executes a bundled script to process local resume files.
Evidence: SKILL.md defines a step to run python scripts/extract_resume.py on user-provided files to extract text and structural metadata.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes content from untrusted external documents.
Ingestion points: scripts/extract_resume.py ingests text from user-uploaded .pdf, .docx, .md, .txt, and .json files.
Boundary markers: The workflow requires the agent to present extracted text to the user for confirmation and modification in Phase 1 before any code generation occurs.
Capability inventory: The skill executes a local Python script for data processing and generates HTML/JS files within the workspace.
Sanitization: The extraction script cleans Markdown formatting and the agent's output is limited by the structured architecture defined in references/HTML_ARCHITECTURE.md.

frontend-resume